Nuclear Engineering and Technology

Korean Nuclear Society (한국원자력학회)
권호 표지
DOI QR코드

DOI QR Code

The ensemble approach in comparison with the diverse feature selection techniques for estimating NPPs parameters using the different learning algorithms of the feed-forward neural network

  • Received : 2021.03.22
  • Accepted : 2021.06.18
  • Published : 2021.12.25
Download PDF
⟨ Previous Next ⟩

Abstract

Several reasons such as no free lunch theorem indicate that there is not a universal Feature selection (FS) technique that outperforms other ones. Moreover, some approaches such as using synthetic dataset, in presence of large number of FS techniques, are very tedious and time consuming task. In this study to tackle the issue of dependency of estimation accuracy on the selected FS technique, a methodology based on the heterogeneous ensemble is proposed. The performance of the major learning algorithms of neural network (i.e. the FFNN-BR, the FFNN-LM) in combination with the diverse FS techniques (i.e. the NCA, the F-test, the Kendall's tau, the Pearson, the Spearman, and the Relief) and different combination techniques of the heterogeneous ensemble (i.e. the Min, the Median, the Arithmetic mean, and the Geometric mean) are considered. The target parameters/transients of Bushehr nuclear power plant (BNPP) are examined as the case study. The results show that the Min combination technique gives the more accurate estimation. Therefore, if the number of FS techniques is m and the number of learning algorithms is n, by the heterogeneous ensemble, the search space for acceptable estimation of the target parameters may be reduced from n × m to n × 1. The proposed methodology gives a simple and practical approach for more reliable and more accurate estimation of the target parameters compared to the methods such as the use of synthetic dataset or trial and error methods.

Keywords

Acknowledgement

The author would like to express his special gratitude to the research office of the Sharif University of Technology for the support in the present study.

References

  1. J.H. Park, Y.J. An, K.H. Yoo, M.G. Na, Leak flow prediction during loss of coolant accidents using deep fuzzy neural networks, Nuclear Engineering and Technology (2021), https://doi.org/10.1016/j.net.2021.01.040.
  2. K. Moshkbar-Bakhshayesh, S. Mohtashami, Classification of NPPs transients using change of representation technique: a hybrid of unsupervised MSOM and supervised SVM, Prog. Nucl. Energy 117 (2019), 103100. https://doi.org/10.1016/j.pnucene.2019.103100
  3. R.E. Uhrig, L.H. Tsoukalas, Soft computing technologies in nuclear engineering applications, Prog. Nucl. Energy 34 (1) (1999) 13-75. https://doi.org/10.1016/S0149-1970(97)00109-1
  4. R.E. Uhrig, J.W. Hines, Computational intelligence in nuclear engineering, Nuclear Engineering and Technology 37 (2) (2005) 127-138.
  5. K. Moshkbar-Bakhshayesh, M.B. Ghofrani, Development of a new method for forecasting future states of NPPs parameters in transients, IEEE Trans. Nucl. Sci. 61 (5) (2014) 2636-2642. https://doi.org/10.1109/TNS.2014.2346234
  6. K. Moshkbar-Bakhshayesh, Prediction of unmeasurable parameters of NPPs using different model-free methods based on cross-correlation detection of measurable/unmeasurable parameters: a comparative study, Ann. Nucl. Energy 139 (2020), 107232. https://doi.org/10.1016/j.anucene.2019.107232
  7. G.P. Choi, K.H. Yoo, J.H. Back, M.G. Na, Estimation of LOCA break size using cascaded fuzzy neural networks, Nuclear Engineering and Technology 49 (3) (2017) 495-503. https://doi.org/10.1016/j.net.2016.11.001
  8. L.V. Fausett, Fundamentals of Neural Networks: Architectures, Algorithms and Applications, Pearson Education India, 2006.
  9. S. Mukhopadhyay, S. Chaudhuri, A feature-based approach to monitor motor-operated valves used in nuclear power plants, IEEE Trans. Nucl. Sci. 42 (6) (1995) 2209-2220. https://doi.org/10.1109/23.489417
  10. B. Xue, M. Zhang, W.N. Browne, Particle swarm optimisation for feature selection in classification: novel initialisation and updating mechanisms, Appl. Soft Comput. 18 (2014) 261-276. https://doi.org/10.1016/j.asoc.2013.09.018
  11. O. Soufan, D. Kleftogiannis, P. Kalnis, V.B. Bajic, DWFS: a wrapper feature selection tool based on a parallel genetic algorithm, PloS One 10 (2) (2015), e0117988. https://doi.org/10.1371/journal.pone.0117988
  12. K. Moshkbar-Bakhshayesh, M. Ghanbari, M.B. Ghofrani, Development of a new features selection algorithm for estimation of NPPs operating parameters, Ann. Nucl. Energy 146 (2020), 107667. https://doi.org/10.1016/j.anucene.2020.107667
  13. D.H. Wolpert, The lack of a priori distinctions between learning algorithms, Neural Comput. 8 (7) (1996) 1341-1390. https://doi.org/10.1162/neco.1996.8.7.1341
  14. V. Bolon-Canedo, A. Alonso-Betanzos, Recent Advances in Ensembles for Feature Selection, Springer, 2018.
  15. C. Sammut, G.I. Webb, Encyclopedia of Machine Learning, Springer Science & Business Media, 2011.
  16. R. Caruana, A. Niculescu-Mizil, An empirical comparison of supervised learning algorithms, in: Proceedings of the 23rd International Conference on Machine Learning, 2006, pp. 161-168.
  17. V. Bolon-Canedo, N. Sanchez-Marono, A. Alonso-Betanzos, Distributed feature selection: an application to microarray data classification, Appl. Soft Comput. 30 (2015) 136-150. https://doi.org/10.1016/j.asoc.2015.01.035
  18. M. Basu, T.K. Ho, Data Complexity in Pattern Recognition, Springer Science & Business Media, 2006.
  19. J.M. Stuart, E. Segal, D. Koller, S.K. Kim, A gene-coexpression network for global discovery of conserved genetic modules, Science 302 (5643) (2003) 249-255. https://doi.org/10.1126/science.1087447
  20. N.S. Chok, Pearson's versus Spearman's and Kendall's Correlation Coefficients for Continuous Data, University of Pittsburgh, 2010.
  21. J. Goldberger, G.E. Hinton, S. Roweis, R.R. Salakhutdinov, Neighbourhood components analysis, Adv. Neural Inf. Process. Syst. 17 (2004) 513-520.
  22. C.A. Markowski, E.P. Markowski, Conditions for the effectiveness of a preliminary test of variance, Am. Statistician 44 (4) (1990) 322-326. https://doi.org/10.2307/2684360
  23. K. Moshkbar-Bakhshayesh, Development of a modular system for estimating attenuation coefficient of gamma radiation: comparative study of different learning algorithms of cascade feed-forward neural network, J. Instrum. 14 (10) (2019). P10010. https://doi.org/10.1088/1748-0221/14/10/p10010
  24. H. Wang, F. Ji, C. Leung, P. Sum, Regularization parameter selection for faulty neural networks, Int. J. Intell. Syst. Technol. 4 (1) (2009) 45-48.
  25. H. Okut, Bayesian Regularized Neural Networks for Small N Big P Data (Artificial Neural Networks-Models Applications), 2016.
  26. K. Moshkbar-Bakhshayesh, Identification of the appropriate architecture of multilayer feed-forward neural network for estimation of NPPs parameters using the GA in combination with the LM and the BR learning algorithms, Ann. Nucl. Energy 156 (2021), 108222. https://doi.org/10.1016/j.anucene.2021.108222
  27. BNPP, Final safety analysis report (FSAR), Chapter 15, Rev. 0,, 2003.